In this piece on Medium, Jenka Gurfinkel writes about a Reddit user who has asked Midjourney, a generative AI to do the following:
Imagine a time traveler journeyed to various times and places throughout human history and showed soldiers and warriors of the periods what a “selfie” is.
The user then tasks Midjourney to fulfil this prompt for ‘Samurai Warriors’, ‘Viking Warriors’, ‘French WW1 Soldiers’, ‘Paleolithic Cavemen’, and so on. Here is what the AI produced for ‘Ancient Native American warriors’:
Gurfinkel notes how all the stereotypically visualised people in these examples have a beaming smile. She is from the former Soviet Union and is well aware that how, when and why we smile is “deeply cultural contextual”. To her these images show how these AIs are trained on datasets from one particular (visual) culture: the American one. And she finds that troubling:
In flattening the diversity of facial expressions of civilizations around the world AI had collapsed the spectrum of history, culture, photography, and emotion concepts into a singular, monolithic perspective. It presented a false visual narrative about the universality of something that in the real world — where real humans have lived and created culture, expression, and meaning for hundreds of thousands of years — is anything but uniform.
See: AI and the American Smile at Medium.